Loading and distribution management and control method and system for electric power measurement materials

Abstract

The invention discloses an electric power measurement material loading and distribution management and control method and system, and belongs to the technical field of data processing, and the method comprises the steps: tracking and monitoring materials through the Internet of Things, an RFID technology and a bar code technology, and generating real-time usage data; querying historical monthly demand data from a preset database; according to the historical monthly demand data and the real-time usage data, using a time sequence analysis and machine learning method to predict future equipment demand; on the basis of deterministic prediction, probability distribution characteristics of electric energy metering material demands are mined, so that a probability distribution function is generated and is used for helping a user to configure inventory more flexibly, and the problem of unreasonable inventory caused by unreal demand information is avoided. According to the invention, the historical inventory and usage data of various types of electric energy metering materials are mined, so that the demand of future equipment can be predicted.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a method and system for controlling the loading and distribution of electric power measurement materials. Background Art

[0002] At present, with the continuous development of information technology, many electric power enterprises have begun to use information systems to control the loading and distribution of electric power measurement materials. These systems usually include a material management system, a distribution management system, etc., and can realize functions such as real-time update of material information and intelligent planning of distribution routes.

[0003] In the actual production management process, many electric power enterprises still adopt relatively traditional and backward management methods in the management of metering devices, and there is still a large gap from the requirements of lean management. The demand for metering equipment in many electric power enterprises is mainly reported by the infrastructure department and high- and low-voltage customer managers. In order to carry out business smoothly, the demand reporters often store a large amount of inventory and increase the reported demand. The inaccuracy of this demand information is likely to cause mistakes in the equipment management of the metering center, or result in capital occupation and unreasonable inventory configuration. Summary of the Invention

[0004] The present application provides a method and system for controlling the loading and distribution of electric power measurement materials, which predict the future equipment demand by mining the historical inventory and usage data of various types of electric energy measurement materials. On the basis of deterministic prediction, the present invention further mines the probability distribution characteristics of the demand for electric energy measurement materials, and further uses the probability distribution characteristics to improve the prediction results. Finally, the prediction result is no longer a single value, but a probability distribution function, and the warehouse manager can flexibly configure the inventory considering the equipment price, equipment importance and warehouse management requirements.

[0005] The method and system for controlling the loading and distribution of electric power measurement materials provided by the present application adopt the following technical solutions:

[0006] In a first aspect, the present application provides a method for controlling the loading and distribution of electric power measurement materials, including the following steps:

[0007] Track and monitor materials through the Internet of Things, RFID technology, and barcode technology to generate real-time usage data;

[0008] Query historical monthly demand data from a preset database;

[0009] Predict the future equipment demand according to the historical monthly demand data and the real-time usage data by using time series analysis and machine learning methods;

[0010] Based on deterministic prediction, mine the probability distribution characteristics of the demand for electrical energy metering materials, so as to generate a probability distribution function, which is used to help users configure the inventory more flexibly and avoid the problem of unreasonable inventory caused by untrue demand information.

[0011] Further, in the step of using time series analysis and machine learning methods to predict the future equipment demand, it specifically includes:

[0012] Preprocess the historical monthly demand data, including removing abnormal data and data normalization, to ensure the quality and consistency of the data;

[0013] Collect and integrate the influencing factor data, where the influencing factor data includes weather, equipment output, attributes, and power grid dispatching data;

[0014] Use a time series analysis model to analyze the trend, cycle, and noise components in the historical monthly demand data and predict future demand;

[0015] Construct a first demand prediction model based on time series to reflect the influence of time factors on demand;

[0016] Construct a neural network model, and use the historical monthly demand data to train the neural network model to capture the dependence relationship between electricity demand and different features;

[0017] Use an LSTM network to predict monthly, quarterly, and annual demands, and improve the prediction accuracy by weighted averaging the monthly, quarterly, and annual prediction results;

[0018] Combine machine learning algorithms to broaden the modeling ability of traditional time series algorithms;

[0019] Based on a time series multi-factor fusion model, incorporate different influencing factors into the prediction model to obtain a second demand prediction model, so as to improve the comprehensiveness and accuracy of the prediction;

[0020] Combine the first demand prediction model and the second demand prediction model, and form a comprehensive prediction model through the principle of maximum information entropy;

[0021] In practical applications, deploy the comprehensive prediction model and conduct tests, compare the differences between the prediction results and the actual demand, verify the effectiveness of the time series prediction model through numerical examples, and adjust the model according to the actual demand.

[0022] Further, in the step of mining the probability distribution characteristics of the demand for electrical energy metering materials based on deterministic prediction to generate a probability distribution function, it also includes:

[0023] Collect and organize historical demand data for various power metering materials, where the historical demand data includes the actual demand quantity per month, as well as the usage frequency and periodic characteristics of the materials;

[0024] Based on the historical demand data, use time series analysis, machine learning models or statistical methods to establish a demand forecasting model for capturing trends and patterns in the historical demand data and predicting future demand;

[0025] Conduct statistical analysis on the historical demand data, calculate mean and variance parameters, and use probability distribution functions to describe the probability characteristics of power material demand. Through the forecasting model, calculate the probability distribution function of material demand within a future period;

[0026] Query the predetermined confidence level and historical data from a preset database;

[0027] Utilize the generated probability distribution function to calculate the safety stock quantity at the predetermined confidence level;

[0028] Set the safety stock level according to the probability distribution of demand to cope with demand fluctuation risks;

[0029] Set inventory strategy parameters according to historical data and business requirements;

[0030] Use simulation software or programming tools to simulate the operation process of the inventory system and record the changes in key indicators during the simulation;

[0031] Analyze the service level and cost - benefit under different inventory strategies and select the optimal strategy;

[0032] Use the CRPS metric to evaluate the accuracy and reliability of probability forecasts, calibrate the model to ensure the consistency of forecast results with actual observations, and optimize the forecasting performance by adjusting model parameters or adopting different kernel functions.

[0033] Furthermore, it also includes using an information system to implement intelligent distribution route planning for power metering materials, specifically including:

[0034] By integrating Internet of Things, big data, cloud computing and artificial intelligence technologies, establish a comprehensive intelligent scheduling platform, which is used to achieve real - time matching of material demand and supply, optimize the matching of vehicles and distribution routes, and improve the overall operation efficiency;

[0035] In terms of route planning, adopt a combination of genetic algorithm and hybrid particle swarm optimization algorithm to improve the operation efficiency and reduce the probability of local optimal solutions, thereby obtaining the best distribution route;

[0036] Use a heuristic orthogonal binary tree search algorithm to rationally allocate metering material vehicles, formulate a distribution plan in combination with actual needs, and optimize the distribution route and loading efficiency;

[0037] Through the Geographic Information System (GIS) and the Global Positioning System (GPS), real-time monitor the location of distribution vehicles, and optimize the location and layout of the distribution center; The GIS is also used to build a geospatial data preprocessing model to provide a scientific basis for route planning;

[0038] Develop an information-based material distribution system based on the ERP system to realize the integration of warehousing management, transportation management and data analysis management; Through the data interface docking with the ERP system, automatically generate logistics orders and realize intelligent distribution management.

[0039] Furthermore, in the aspect of path planning, in the steps of adopting a combination of genetic algorithm and hybrid particle swarm optimization algorithm to improve the operation efficiency, it specifically includes:

[0040] Use the triangle method to model the spatial environment and generate the basic data for initial path planning;

[0041] Implement the artificial potential field method in the genetic algorithm to design the initial path planning;

[0042] Set the target point as the attractive potential and the obstacle as the repulsive point, and guide the moving object to approach the target position through the virtual force field;

[0043] Adopt a parallel search method to seek the global optimal solution, combine the genetic algorithm and the artificial potential field method, and use the artificial potential field method to optimize the global path obtained by the genetic algorithm;

[0044] Combine the advantages of the genetic algorithm and the particle swarm optimization algorithm to form a hybrid algorithm; Use the initial path generated by the genetic algorithm as the input of the particle swarm optimization algorithm, and further optimize the path through the particle swarm optimization algorithm; The particle swarm optimization algorithm adjusts the position and velocity of the particles, and uses the global search ability and local search ability to quickly converge to the optimal solution;

[0045] Use the cubic spline interpolation method to smooth the path generated by the hybrid algorithm and reduce the collision risk in the path;

[0046] Verify the effectiveness of the proposed algorithm through simulation experiments.

[0047] Furthermore, it also includes establishing a material demand plan and a safety inventory quota system, and optimizing the procurement strategy through data analysis to reduce inventory backlog and capital occupation, specifically including:

[0048] Apply machine learning models to conduct in-depth analysis of material usage, identify demand fluctuations, seasonal factors, and market trends, and formulate demand forecasting models;

[0049] Through big data analysis techniques, analyze historical sales data, market trends, and consumer behavior to predict future material demands;

[0050] Based on the big data analysis results, develop an intelligent decision support system to automatically generate material demand plans, procurement suggestions, and inventory adjustment plans according to preset rules and algorithms;

[0051] Monitor inventory levels and sales speeds in real time, identify the risks of overstocking or out-of-stock, and adjust replenishment strategies based on real-time data to ensure sufficient inventory without waste;

[0052] Based on the big data analysis results, adjust the safety inventory level to reduce inventory costs and reduce capital occupancy; by analyzing inventory turnover rate and out-of-stock rate indicators, identify slow-moving, overstocked, and shortage materials, and formulate corresponding adjustment strategies.

[0053] Furthermore, it also includes:

[0054] Query key indicators and evaluation criteria from a preset database. The key indicators include the quality, delivery punctuality, price competitiveness, and supply capacity of suppliers; the evaluation criteria are pre-established by users according to the specific needs and characteristics of the enterprise;

[0055] Establish a supplier database and collect basic information of suppliers. The basic information of suppliers includes company qualifications, scale, experience, and performance;

[0056] Conduct multi-dimensional comprehensive evaluations of suppliers, regularly evaluate the performance of suppliers, and select suppliers with high stability and reliability;

[0057] Regularly review and adjust evaluation indicators and criteria to ensure they always meet the actual needs of the enterprise.

[0058] On the second aspect, the present application provides a loading and distribution control system for power metering materials, including:

[0059] A real-time usage data generation module, which is used to track and monitor materials through Internet of Things, RFID technology, and barcode technology to generate real-time usage data;

[0060] A historical monthly demand data query module, which is used to query historical monthly demand data from a preset database;

[0061] An equipment demand forecasting module, which is used to predict future equipment demands according to the historical monthly demand data and real-time usage data by using time series analysis and machine learning methods;

[0062] A probability distribution function generation module is used to mine the probability distribution characteristics of the power metering material demand on the basis of deterministic prediction, so as to generate a probability distribution function, which helps users configure the inventory more flexibly and avoid the unreasonable inventory problems caused by untrue demand information.

[0063] In a third aspect, the present application provides an intelligent terminal, including a memory and a processor, and a computer program capable of being loaded and executed by the processor for the above-mentioned method for controlling the loading and distribution of power metering materials is stored on the memory.

[0064] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program capable of being loaded and executed by the processor for the above-mentioned method for controlling the loading and distribution of power metering materials.

[0065] In summary, compared with the prior art, the beneficial effects of the above technical solutions are as follows:

[0066] For the method and system for controlling the loading and distribution of power metering materials according to the present application, by using the Internet of Things, RFID technology and bar code technology, real-time tracking and monitoring of power metering materials can be realized. This not only improves the transparency and accuracy of material management, but also ensures the timeliness of data. Managers can immediately understand the current status and location of materials, providing reliable data support for subsequent decision-making. By querying historical monthly demand data from a preset database and combining it with real-time usage data, the demand situation of materials can be understood more comprehensively. This method of combining historical and real-time data analysis provides a more accurate and comprehensive data basis for future demand prediction.

[0067] Using time series analysis and machine learning methods to predict future equipment demand can greatly improve the accuracy of prediction. Time series analysis can capture the trends and periodicities of data changes over time, while machine learning can handle more complex data relationships, thus obtaining more accurate prediction results. On the basis of deterministic prediction, further mine the probability distribution characteristics of the power metering material demand and generate a probability distribution function. This step enables users to configure the inventory more flexibly and avoid unreasonable inventory problems caused by inaccurate demand information or demand fluctuations. The probability distribution function provides a range of possibilities for inventory configuration, enabling users to make more reasonable decisions according to the actual situation and risk preference. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a schematic flowchart of a method for controlling the loading and distribution of power metering materials according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] The present application will be further described in detail below in conjunction with all the accompanying drawings.

[0070] An embodiment of the present application discloses a method and system for controlling the loading and distribution of electric power metering materials. Referring to Figure 1 , a method for controlling the loading and distribution of electric power metering materials includes:

[0071] S101. Generate real-time usage data.

[0072] Specifically, the control system tracks and monitors the materials through the Internet of Things, RFID technology, and barcode technology to generate real-time usage data; real-time collects the status information of electric power metering materials, such as location, quantity, status, etc., through devices such as Internet of Things sensors, and uploads this information to the management system. Attach an RFID tag to each electric power metering material, and read the tag information through an RFID reader to achieve the tracking and monitoring of the materials. At the same time, RFID technology can also realize the automatic identification and information collection of materials, improving management efficiency. Generate a unique barcode for the electric power metering materials, and the basic information and usage data of the materials can be obtained by scanning the barcode. Real-time track and monitor the status and location of the materials to ensure the accuracy and safety of the materials, automatically collect and upload the material information, reduce manual operations, improve work efficiency, generate real-time usage data, and provide data support for subsequent demand forecasting and inventory allocation.

[0073] S102. Query historical monthly demand data from a preset database.

[0074] Specifically, the control system queries historical monthly demand data from a preset database; establishes a database containing historical monthly demand data, and the data should cover the demand for electric power metering materials over a period of time in the past. Extract the corresponding historical monthly demand data from the database according to the time period and material type to be queried. It is beneficial to provide comprehensive historical data support, provide a reliable data basis for subsequent demand forecasting, facilitate the analysis of the change trend and periodicity of material demand, and provide a basis for subsequent management decisions.

[0075] S103. Use time series analysis and machine learning methods to predict future equipment demand.

[0076] Specifically, the control system uses time series analysis and machine learning methods to predict future equipment demand based on the historical monthly demand data and real-time usage data. It performs time series analysis on the historical monthly demand data to identify trends, seasonality, and cyclical components in the data, and builds a prediction model based on these components. Machine learning algorithms (such as linear regression, support vector regression, decision tree, random forest, etc.) are used to train and learn from the historical monthly demand data and real-time usage data to build a prediction model. By continuously adjusting model parameters and optimizing algorithms, the accuracy of the prediction is improved. Thus, the future equipment demand can be accurately predicted, providing a scientific basis for subsequent inventory allocation and loading and distribution, improving the accuracy of the prediction, and reducing inventory backlogs or shortages caused by inaccurate predictions.

[0077] S104. Mine the probability distribution characteristics of the demand for electrical energy measurement materials, thereby generating a probability distribution function.

[0078] Specifically, based on the deterministic prediction, the control system mines the probability distribution characteristics of the demand for electrical energy measurement materials, thereby generating a probability distribution function to help users configure inventory more flexibly and avoid unreasonable inventory problems caused by untrue demand information. Based on the deterministic prediction, the prediction results are further analyzed and processed to mine the probability distribution characteristics of the demand for electrical energy measurement materials. According to the probability distribution characteristics, a probability distribution function is generated to describe the possibility and distribution of the material demand. A more flexible and scientific inventory allocation plan is provided to avoid unreasonable inventory problems. According to the probability distribution function, users can configure inventory more flexibly to cope with demand uncertainty, reduce inventory costs, improve inventory turnover, and improve the economic benefits of the enterprise.

[0079] In another embodiment, S103 specifically includes the following sub-steps:

[0080] S103.1. Preprocess the historical monthly demand data.

[0081] Specifically, the control system preprocesses the historical monthly demand data, including removing abnormal data and data normalization, to ensure the quality and consistency of the data; checks the historical monthly demand data, identifies and removes abnormal data points caused by equipment failures, data entry errors, etc. Standardize the historical monthly demand data and convert it to the same dimension for subsequent model training and analysis. Improve data quality, ensure the accuracy and stability of the prediction model, eliminate the impact of different data dimensions on the prediction results, and improve the generalization ability of the model.

[0082] S103.2. Collect and integrate influencing factor data.

[0083] Specifically, the control system collects and integrates the influencing factor data, which includes weather, equipment output, attributes, and power grid dispatching data; obtains data such as weather, equipment output, attributes, and power grid dispatching from relevant data sources (such as meteorological departments, power grid dispatching centers, etc.). Clean and organize the collected data, and match it with the historical monthly demand data to form a complete data set. Introduce more influencing factors to improve the comprehensiveness and accuracy of the prediction model, provide rich feature information for the model, and help capture the dependence relationship between power demand and different features.

[0084] S103.3. Analyze the trend, cycle, and noise components in the historical monthly demand data using a time series analysis model to predict future demand.

[0085] Specifically, the control system can predict future data and visualize and interpret the results to guide decision-making by selecting and optimizing the model. Use the trained model to predict future data to obtain the predicted value and its confidence interval. Visualize the prediction results, such as using line charts, bar charts, etc., to intuitively observe the prediction trend and periodic changes. Conduct in-depth analysis of the prediction results to explain the differences between the predicted value and the actual value and their possible reasons. At the same time, adjust and optimize the decision-making strategy according to the prediction results.

[0086] Thereby identify the long-term trend in the historical monthly demand data, such as an increasing or decreasing trend. Identify the periodic components in the data, such as seasonal fluctuations; separate the random noise in the data to improve the accuracy of the prediction. Reveal the long-term trend and periodic pattern of power demand, provide a scientific basis for prediction, reduce the interference of noise on the prediction results, and improve the reliability of the prediction.

[0087] S103.4. Construct a first demand prediction model based on time series.

[0088] Specifically, the control system constructs a first demand prediction model based on time series to reflect the impact of time factors on demand; select a suitable time series analysis model (such as ARIMA, exponential smoothing, etc.) according to the characteristics of the historical monthly demand data. Use the historical monthly demand data to train the model and determine the model parameters. Reflect the impact of time factors on power demand, provide preliminary prediction results, and lay a foundation for the subsequent comprehensive prediction model.

[0089] S103.5. Construct a neural network model.

[0090] Specifically, the control system constructs a neural network model and trains the neural network model using the historical monthly demand data to capture the dependency relationships between electricity demand and different features. According to the characteristics of the historical monthly demand data and the influencing factor data, a suitable neural network structure (such as a multi-layer perceptron, a convolutional neural network, etc.) is designed. The neural network model is trained using the historical monthly demand data and the influencing factor data to capture the dependency relationships between electricity demand and different features. Furthermore, complex non-linear relationships are captured, the prediction accuracy and precision are improved, and a powerful supplement is provided for the comprehensive prediction model.

[0091] S103.6. Use an LSTM network to predict monthly, quarterly, and annual demands.

[0092] Specifically, the control system uses a long short-term memory network (LSTM) to construct a prediction model to capture the long-term dependency relationships in time series data. The LSTM network is used to predict monthly, quarterly, and annual demands respectively. The monthly, quarterly, and annual prediction results are weighted and averaged to improve the prediction accuracy. The long-term dependency relationships in time series data are captured, the prediction stability is improved, and through weighted averaging, the volatility of a single prediction result is reduced and the prediction reliability is improved.

[0093] The dataset is divided into a training set, a validation set, and a test set. An LSTM neural network model is constructed, including an input layer, an LSTM layer, and an output layer. According to the demand and data characteristics, hyperparameters such as the number of LSTM layers and the number of neurons in each layer are determined. An activation function and an optimizer are selected to optimize the model parameters. The training set is used to train the LSTM model, and indicators such as the loss function value and the accuracy rate during the training process are recorded. The trained LSTM model is evaluated using the test set, and indicators such as the prediction error and the accuracy rate are calculated. According to the evaluation results, the generalization ability and prediction performance of the model are judged. The trained LSTM model is used to predict future monthly, quarterly, and annual demands. The prediction results are de-normalized or de-standardized to obtain the actual demand prediction values.

[0094] S103.7. Broaden the modeling ability of traditional time series algorithms.

[0095] Specifically, the control system combines machine learning algorithms to broaden the modeling ability of traditional time series algorithms; according to the data characteristics and prediction requirements, a suitable machine learning algorithm (such as a support vector machine, a random forest, etc.) is selected. The machine learning algorithm is combined with the traditional time series algorithm to broaden the modeling ability. By combining the advantages of different algorithms, the comprehensiveness and accuracy of the prediction are improved, and more diverse modeling means are provided for the comprehensive prediction model.

[0096] S103.8. Based on a time series multi-factor fusion model, different influencing factors are incorporated into the prediction model to obtain a second demand prediction model.

[0097] Specifically, based on the time - series multi - factor fusion model, the control system incorporates different influencing factors into the prediction model to obtain the second demand prediction model, so as to improve the comprehensiveness and accuracy of the prediction; incorporates different influencing factors (such as weather, equipment output, etc.) into the time - series analysis model to construct a multi - factor fusion model. The multi - factor fusion model is trained using historical monthly demand data and influencing factor data. Considering more influencing factors, it improves the comprehensiveness and accuracy of the prediction model and provides richer feature information for the comprehensive prediction model.

[0098] S103.9. Combine the first demand prediction model and the second demand prediction model, and form a comprehensive prediction model through the principle of maximum information entropy.

[0099] Specifically, the control system combines the first demand prediction model and the second demand prediction model, and forms a comprehensive prediction model through the principle of maximum information entropy; uses the principle of maximum information entropy to combine the first demand prediction model and the second demand prediction model to form a comprehensive prediction model. By synthesizing the prediction results of different models, it improves the accuracy and robustness of the prediction, providing a reliable prediction basis for subsequent practical applications.

[0100] S103.10. Deploy the comprehensive prediction model and conduct tests to compare the differences between the prediction results and the actual demands.

[0101] Specifically, in actual applications, the control system deploys the comprehensive prediction model and conducts tests to compare the differences between the prediction results and the actual demands, verifies the effectiveness of the time - series prediction model through numerical examples, and adjusts the model according to the actual demands. The comprehensive prediction model is deployed to the actual application scenario. Tests are carried out using actual data, and the differences between the prediction results and the actual demands are compared. The effectiveness of the time - series prediction model is verified through numerical examples. The model parameters and structure are adjusted according to the actual demands to improve the adaptability and accuracy of the model. The accuracy and effectiveness of the model are verified, providing a scientific basis for subsequent practical applications, and the model is adjusted according to the actual demands to improve the adaptability and practicality of the model.

[0102] In another embodiment, S104 specifically includes the following sub - steps:

[0103] S104.1. Collect and organize the historical demand data of various electric energy measurement materials.

[0104] Specifically, the control system collects and organizes the historical demand data of various electric energy measurement materials. The historical demand data includes the actual demand quantity per month, as well as the usage frequency and periodic characteristics of the materials; extract the historical demand data of various electric energy measurement materials from the enterprise database, ERP system or logistics system, including the actual demand quantity per month, the usage frequency and periodic characteristics of the materials, etc. Clean and organize the data to remove outliers and duplicate data to ensure the accuracy and consistency of the data. Provide a reliable data basis for subsequent demand forecasting and probability distribution function generation, reveal the periodic characteristics and trends of material demand, and help formulate more reasonable inventory strategies.

[0105] S104.2. Use time series analysis, machine learning models or statistical methods to establish a demand forecasting model.

[0106] Specifically, based on the historical demand data, the control system uses time series analysis, machine learning models or statistical methods to establish a demand forecasting model to capture the trends and patterns in the historical demand data and predict future demands; based on the collected historical demand data, use time series analysis (such as ARIMA, exponential smoothing, etc.), machine learning models (such as neural networks, support vector machines, etc.) or statistical methods (such as regression analysis, distribution fitting, etc.) to establish a demand forecasting model. Train and validate the model to ensure that it can accurately capture the trends and patterns in the historical demand data and predict future demands. Furthermore, it can provide a preliminary prediction result of material demand for a period of time in the future, providing a scientific basis for subsequent probability distribution function generation and inventory strategy formulation.

[0107] S104.3. Conduct statistical analysis on the historical demand data, calculate the mean and variance parameters, and use a probability distribution function to describe the probability characteristics of the power material demand.

[0108] Specifically, the control system conducts statistical analysis on the historical demand data, calculates the mean and variance parameters, and uses a probability distribution function to describe the probability characteristics of the power material demand. Through the forecasting model, calculate the probability distribution function of material demand for a period of time in the future; conduct statistical analysis on the historical demand data and calculate statistical parameters such as the mean and variance. According to the statistical parameters and the distribution of the data, select an appropriate probability distribution function (such as normal distribution, Poisson distribution, exponential distribution, etc.) to describe the probability characteristics of the power material demand. Through the forecasting model, calculate the probability distribution function of material demand for a period of time in the future. Reveal the probability distribution characteristics of material demand, provide a basis for subsequent safety stock inventory calculation and inventory strategy formulation, improve the flexibility and accuracy of inventory management, and reduce inventory costs.

[0109] S104.4. Query the predetermined confidence level and historical data from the preset database.

[0110] Specifically, the control system queries the predetermined confidence level and historical data from a preset database; queries the predetermined confidence level (such as 95%, 99%, etc.) set by the enterprise and historical demand data from the preset database, and calculates the threshold of the safety stock quantity according to the predetermined confidence level and historical data. Ensure that the inventory can meet the material requirements within a certain period in the future at the predetermined confidence level, and improve the reliability and stability of inventory management.

[0111] S104.5. Calculate the safety stock quantity at the predetermined confidence level using the generated probability distribution function.

[0112] Specifically, the control system uses the generated probability distribution function to calculate the safety stock quantity at the predetermined confidence level; determines the probability distribution type of inventory demand based on information such as historical sales data and market demand. Common probability distribution types include normal distribution, Poisson distribution, etc. If the data shows continuous and symmetric distribution characteristics, the normal distribution can be considered; if the data represents the number of occurrences of random events within a certain period, the Poisson distribution is more appropriate. Collect historical sales data, market demand data, etc. of the product to calculate the parameters of the probability distribution function.

[0113] For the normal distribution, the mean and standard deviation need to be calculated; for the Poisson distribution, the average number of occurrences λ per unit time needs to be calculated. Generate the corresponding probability distribution function according to the determined probability distribution type and the calculated parameters. For example, for the Poisson distribution, its probability distribution function is P(X = k) = (λ^(-λ)) / k!, where X represents the number of occurrences of the event, k represents the specific number of occurrences of the event, and λ is the average number of occurrences per unit time. Use the generated probability distribution function and the set confidence level to calculate the safety stock quantity. The specific method varies according to the different probability distribution types. For example, for the Poisson distribution, the value of k that satisfies the predetermined confidence level can be solved, and then the safety stock quantity is calculated as k - λ.

[0114] Calculating the safety stock quantity through scientific methods avoids excessive or insufficient inventory backlogs and improves inventory turnover. Reduces the shortage cost caused by insufficient inventory and the holding cost caused by excessive inventory. Sets a predetermined confidence level to ensure that the inventory quantity can meet the demand under a certain probability and reduces the shortage risk. At the same time, by real-time monitoring the changes in inventory levels and market demand, the inventory strategy can be adjusted in a timely manner to further reduce risks.

[0115] Reasonably allocate resources such as procurement, production, and logistics according to market demand and inventory conditions. Improves the utilization efficiency of resources and reduces operating costs. By ensuring sufficient inventory quantity to meet customer needs, customer satisfaction and loyalty are improved. Reduces customer complaints and losses caused by shortages.

[0116] S104.6. Set the safety inventory level according to the probability distribution of demand.

[0117] Specifically, the control system sets the safety inventory level according to the probability distribution of demand to cope with the risk of demand fluctuations; uses the generated probability distribution function and the predetermined confidence level to calculate the safety inventory quantity. According to the calculation results, a reasonable safety inventory level is set to cope with the risk of demand fluctuations. Ensure that when demand fluctuates, the inventory can maintain a certain buffer capacity, improving the flexibility and response speed of inventory management.

[0118] S104.7. Set the inventory policy parameters according to historical data and business requirements.

[0119] Specifically, the control system sets the inventory policy parameters, such as reorder point, order quantity, etc., according to historical data and business requirements. Ensure that the inventory policy can meet the actual needs of the enterprise and reduce inventory costs. It is beneficial to improve the efficiency and accuracy of inventory management, optimize the inventory structure, and reduce the risks of inventory backlog and stockout.

[0120] S104.8. Use simulation software or programming tools to simulate the operation process of the inventory system and record the changes in key indicators during the simulation process.

[0121] Specifically, the control system uses simulation software or programming tools to simulate the operation process of the inventory system according to the set inventory policy parameters and probability distribution function. Record the changes in key indicators during the simulation process, such as inventory level, stockout rate, order frequency, etc. Evaluate the system performance under different inventory policies to provide data support for the optimization of inventory policies.

[0122] S104.9. Analyze the service level and cost - benefit under different inventory policies.

[0123] Specifically, the control system analyzes the service level and cost - benefit under different inventory policies according to the simulation results. Compare the advantages and disadvantages of different policies and select the optimal policy. Improve the economic and social benefits of inventory management, optimize the inventory policy, and reduce inventory costs.

[0124] S104.10. Use the CRPS index to evaluate the accuracy and reliability of probability forecasting.

[0125] Specifically, the control system uses the CRPS metric to evaluate the accuracy and reliability of probability forecasts, calibrates the model to ensure the consistency of the forecast results with the actual observations, and optimizes the forecast performance by adjusting the model parameters or adopting different kernel functions. The Continuous Ranked Probability Score (CRPS) metric is used to evaluate the accuracy and reliability of probability forecasts. Based on the calculation results of the CRPS metric, the model is calibrated to ensure the consistency of the forecast results with the actual observations. The forecast performance is optimized by adjusting the model parameters or adopting different kernel functions. Improve the accuracy and reliability of probability forecasts, optimize the forecast model, and enhance the scientific nature and accuracy of inventory management.

[0126] In another embodiment, an information system is also used to implement intelligent distribution route planning for electric power measurement materials, which specifically includes the following sub-steps:

[0127] S201. Establish a comprehensive intelligent scheduling platform.

[0128] Specifically, the control system integrates Internet of Things, big data, cloud computing, and artificial intelligence technologies to establish a comprehensive intelligent scheduling platform. The intelligent scheduling platform is used to achieve real-time matching of material demand and supply, optimize the matching of vehicles and distribution routes, and improve the overall operation efficiency. By integrating Internet of Things (IoT), big data, cloud computing, and artificial intelligence technologies, a comprehensive intelligent scheduling platform is constructed. This platform can collect and analyze multi-dimensional data such as material demand, supply situation, vehicle status, and road conditions in real time, and use algorithm models for intelligent decision-making to achieve real-time matching of material demand and supply, as well as optimized matching of vehicles and distribution routes. Significantly improve the accuracy and timeliness of material distribution, optimize resource allocation, reduce vehicle empty running and waiting time, and improve the overall operation efficiency.

[0129] S202. Adopt a combination of genetic algorithm and hybrid particle swarm optimization algorithm to obtain the best distribution route.

[0130] Specifically, in terms of route planning, the control system adopts a combination of genetic algorithm (GA) and hybrid particle swarm optimization algorithm (PSO). The genetic algorithm searches for the optimal solution by simulating the biological evolution process, while the particle swarm optimization algorithm finds the optimal path by simulating the group behavior of bird flocks or fish schools. Combining the two can give full play to their respective advantages, improve the operation efficiency, and reduce the risk of falling into local optimal solutions, thereby obtaining a global optimal or approximate global optimal distribution route. Improve the accuracy and efficiency of route planning, reduce distribution costs, and enhance customer satisfaction.

[0131] S203. Use the heuristic orthogonal binary tree search algorithm to reasonably allocate metering material vehicles and formulate a distribution plan in combination with actual requirements.

[0132] Specifically, the heuristic orthogonal binary tree search algorithm combines the advantages of heuristic search and binary tree structure. It guides the search process through an evaluation function, preferentially searching for heuristic nodes, thereby quickly finding the optimal or near-optimal delivery route. Each delivery point (including the starting point, customer points, and ending point) is defined as a node of the binary tree, and the node contains location information, demand information, etc. According to the delivery objectives (such as cost, time, etc.), an evaluation function is designed to evaluate the advantages and disadvantages of each node. The evaluation function can comprehensively consider multiple factors such as distance, time, and cost. Starting from the starting node, the evaluation function is used to guide the search process, preferentially searching for nodes with smaller evaluation values. During the search process, new nodes are continuously generated and the evaluation values are updated until the target node is found or the termination condition is met. The optimal or near-optimal delivery route and the corresponding vehicle configuration plan are output.

[0133] According to the results of the heuristic orthogonal binary tree search algorithm, determine the delivery route and loading capacity of each vehicle to ensure that the vehicle can meet the delivery requirements without overloading. Combining the actual situation (such as road conditions, traffic rules, etc.), further optimize the delivery route to ensure the feasibility and efficiency of the route. Evaluate the formulated delivery plan, including aspects such as cost, time, and service quality, to ensure that the plan meets the overall strategy of the enterprise and customer needs. The control system uses the heuristic orthogonal binary tree search algorithm to reasonably allocate metering material vehicles. This algorithm searches and evaluates possible delivery plans by constructing a binary tree structure, and finds the delivery plan that meets the actual needs and has the lowest cost. At the same time, formulate a delivery plan in combination with the actual needs, optimize the delivery path and loading efficiency. Realize the reasonable allocation and efficient utilization of vehicle resources, and improve the flexibility and adaptability of the delivery path.

[0134] S204. Real-time monitor the location of the delivery vehicle through the Geographic Information System (GIS) and the Global Positioning System (GPS).

[0135] Specifically, the control system real-time monitors the location of the delivery vehicle through the Geographic Information System (GIS) and the Global Positioning System (GPS), and optimizes the location selection and layout of the distribution center; the GIS is also used to construct a geospatial data preprocessing model to provide a scientific basis for route planning; real-time monitor the location and status of the delivery vehicle through the Geographic Information System (GIS) and the Global Positioning System (GPS). The GIS is used to construct a geospatial data preprocessing model to provide a scientific basis for route planning; the GPS is used to real-time track the vehicle location to ensure the safety and accuracy of the delivery process. At the same time, use the GIS to optimize the location selection and layout of the distribution center, reduce transportation costs, and improve delivery efficiency. Realize the real-time monitoring and dynamic adjustment of the delivery process, optimize the distribution center layout, and improve the overall delivery efficiency.

[0136] S205. Develop an information-based material distribution system based on the ERP system to achieve the integration of warehousing management, transportation management, and data analysis management.

[0137] Specifically, the control system develops an information-based material distribution system based on the ERP system to achieve the integration of warehousing management, transportation management, and data analysis management; through the data interface docking with the ERP system, logistics orders are automatically generated to achieve intelligent distribution management. Develop an information-based material distribution system based on the Enterprise Resource Planning (ERP) system. This system realizes the integration of warehousing management, transportation management, and data analysis management. Through the data interface docking with the ERP system, logistics orders can be automatically generated to achieve intelligent distribution management. At the same time, the system can also conduct in-depth analysis of distribution data to provide data support for future distribution decisions. Realize the collaborative management of warehousing, transportation, and data analysis, and improve the intelligence and automation level of distribution management.

[0138] In another embodiment, S202 specifically includes the following sub-steps:

[0139] S202.1. Use the triangle method to model the space environment and generate the basic data for initial path planning.

[0140] Specifically, the control system uses the triangle method to model the space environment and generate the basic data for initial path planning; uses the triangle method to divide the space environment into multiple triangular regions, and each region has specific attributes and characteristics. These attributes can include terrain, obstacle positions, target point positions, etc. Through this method, the basic data for initial path planning can be generated. The space environment is effectively simplified and abstracted, providing a clear and easy-to-process model for subsequent path planning.

[0141] S202.2. Implement the artificial potential field method in the genetic algorithm to design the initial path planning.

[0142] Specifically, the control system implements the artificial potential field method in the genetic algorithm to design the initial path planning; embeds the artificial potential field method in the genetic algorithm to design the initial path planning. The artificial potential field method establishes an artificial potential field in space and uses attractive and repulsive points to guide the moving object to approach the target position and avoid obstacles. Combining the global search ability of the genetic algorithm and the local guiding ability of the artificial potential field method provides an effective solution for the initial path planning.

[0143] S202.3. Set the target point as the attractive potential and the obstacle as the repulsive point, and guide the moving object to approach the target position through the virtual force field.

[0144] Specifically, the control system sets the target point as an attractive potential and the obstacle as a repulsive point. By calculating the virtual forces of each potential point on the moving object, the moving object can be guided to approach the target position along the optimal path. Under the guidance of the virtual force field, the moving object can efficiently avoid obstacles and approach the target point, improving the safety and accuracy of path planning.

[0145] S202.4. Adopt a parallel search method to find the global optimal solution, combine the genetic algorithm and the artificial potential field method, and use the artificial potential field method to optimize the global path obtained by the genetic algorithm.

[0146] Specifically, the control system combines the genetic algorithm and the artificial potential field method, and uses a parallel search method to find the global optimal solution. In the genetic algorithm, better solutions are continuously evolved through operations such as selection, crossover, and mutation; at the same time, the artificial potential field method is used to optimize the global path obtained by the genetic algorithm. The parallel search method improves the search efficiency of the algorithm and can find the global optimal solution or an approximate optimal solution in a short time.

[0147] S202.5. Combine the advantages of the genetic algorithm and the particle swarm optimization algorithm to form a hybrid algorithm; use the initial path generated by the genetic algorithm as the input of the particle swarm optimization algorithm, and further optimize the path through the particle swarm optimization algorithm.

[0148] Specifically, the particle swarm optimization algorithm adjusts the position and velocity of the particles, utilizes the global search ability and the local search ability, and quickly converges to the optimal solution; combines the advantages of the genetic algorithm and the particle swarm optimization algorithm to form a hybrid algorithm. Use the initial path generated by the genetic algorithm as the input of the particle swarm optimization algorithm, and further optimize the path through the particle swarm optimization algorithm. The particle swarm optimization algorithm adjusts the position and velocity of the particles and quickly converges to the optimal solution by utilizing the global search ability and the local search ability. The hybrid algorithm combines the global search ability of the genetic algorithm and the fast convergence ability of the particle swarm optimization algorithm, and can efficiently plan the optimal path.

[0149] S202.6. Use the cubic spline interpolation method to smooth the path generated by the hybrid algorithm.

[0150] Specifically, the control system uses the cubic spline interpolation method to smooth the path generated by the hybrid algorithm. Through interpolation calculation, a continuous and smooth path can be obtained, reducing the collision risk in the path. The smoothed path is more in line with the motion characteristics of the actual moving object, reducing the risks of collision and bump, and improving the safety and feasibility of path planning.

[0151] S202.7. Verify the effectiveness of the proposed algorithm through simulation experiments.

[0152] Specifically, the control system verifies the effectiveness of the proposed algorithm through simulation experiments. Design simulation experiments and apply the proposed algorithm to the actual path planning problem. Evaluate the effectiveness of the algorithm by comparing the output results of the algorithm with the expected results. The simulation experiment results show that the proposed algorithm can efficiently plan the optimal path, and the path is smooth, safe, and feasible. This verifies the effectiveness and practicality of the algorithm.

[0153] Furthermore, as another implementation manner, the embodiments of the present application can also establish a material demand plan and a safety inventory quota system, and optimize the procurement strategy through data analysis to reduce inventory backlog and capital occupation, including the following steps:

[0154] S301. Apply a machine learning model to deeply analyze the material usage situation.

[0155] Specifically, the control system applies a machine learning model to deeply analyze the material usage situation, identify demand fluctuations, seasonal factors, and market trends, and formulate a demand prediction model; collect historical material usage data, including usage quantity, usage time, usage frequency, etc., and apply a machine learning model, such as time series analysis, regression model, etc., to deeply analyze the material usage situation. Identify demand fluctuations, seasonal factors, and market trends, and formulate a demand prediction model. According to the prediction results, formulate a material demand plan. Improve the accuracy of material demand prediction, reduce inventory backlog or out-of-stock phenomena caused by inaccurate prediction, provide data support for the formulation of procurement strategies, and improve procurement efficiency.

[0156] S302. Analyze historical sales data, market trends, and consumer behavior to predict future material demands.

[0157] Specifically, the control system analyzes historical sales data, market trends, and consumer behavior through big data analysis technology to predict future material demands; collect historical sales data, market trends, and consumer behavior data. Use big data analysis technology, such as data mining, clustering analysis, etc., to deeply analyze the data. According to the analysis results, predict future material demands. According to the prediction results, adjust the material demand plan and procurement strategy. Improve the prediction ability of future material demands, provide a scientific basis for the formulation of procurement strategies, and reduce inventory backlog and capital occupation caused by inaccurate prediction.

[0158] S303. Develop an intelligent decision support system according to the big data analysis results, and automatically generate a material demand plan, procurement suggestions, and inventory adjustment plans according to preset rules and algorithms.

[0159] Specifically, the control system develops an intelligent decision support system based on the big data analysis results. The system automatically generates material requirement plans, procurement suggestions, and inventory adjustment plans according to preset rules and algorithms. The system is updated and optimized regularly to improve its accuracy and reliability. This enhances the efficiency and accuracy of decision-making, reduces decision-making errors caused by human factors, provides intelligent support for procurement and inventory management, and reduces operating costs.

[0160] S304. Monitor the inventory level and sales speed in real time, identify the risks of overstocking or out-of-stock, and adjust the replenishment strategy based on real-time data.

[0161] Specifically, the control system establishes a real-time inventory monitoring system to monitor the inventory level and sales speed. Set inventory warning thresholds. When the inventory level is below or above the threshold, the system automatically issues a warning. Adjust the replenishment strategy according to real-time data to ensure sufficient inventory without waste. This helps to promptly detect the risks of overstocking or out-of-stock, avoid inventory backlogs and out-of-stock situations, improve inventory turnover, and reduce inventory costs.

[0162] S305. Adjust the safety inventory level according to the big data analysis results.

[0163] Specifically, the control system adjusts the safety inventory level according to the big data analysis results to reduce inventory costs and capital occupation; by analyzing inventory turnover rate, out-of-stock rate and other indicators, identify slow-moving, overstocked and shortage materials, and formulate corresponding adjustment strategies. Adjust the safety inventory level according to the big data analysis results. Set reasonable safety inventory thresholds to ensure that material requirements can be met in case of emergencies. Regularly evaluate and adjust the safety inventory level to adapt to market changes. Reduce inventory costs and capital occupation, and improve the response speed and flexibility of the supply chain.

[0164] By analyzing indicators such as inventory turnover rate and out-of-stock rate, identify slow-moving, overstocked and shortage materials, and formulate corresponding adjustment strategies such as promotions, returns, replenishments, etc. Regularly evaluate and optimize the effectiveness of the adjustment strategies. Reduce the quantity of slow-moving and overstocked materials, reduce inventory costs, ensure the timely supply of shortage materials, and improve customer satisfaction.

[0165] Furthermore, as another implementation method, the embodiments of the present application may further include the following steps:

[0166] S401. Query key indicators and evaluation criteria from a preset database

[0167] Specifically, the control system queries key indicators and evaluation criteria from a preset database. The key indicators include the quality, delivery punctuality, price competitiveness, and supply capacity of suppliers. The evaluation criteria are pre-established by users according to the specific needs and characteristics of the enterprise. The enterprise sets and stores key indicators in the preset database according to its own needs, such as the quality, delivery punctuality, price competitiveness, supply capacity, etc. of suppliers. The evaluation criteria are pre-established by users according to the specific needs and characteristics of the enterprise and are also stored in the database. These criteria include specific values of quality standards (such as product qualification rate, defective rate, etc.), thresholds for delivery punctuality, comparison benchmarks for price competitiveness, and evaluation methods for supply capacity, etc. When conducting supplier evaluations, these key indicators and evaluation criteria are queried from the database and used as the basis for evaluation. This ensures the consistency and objectivity of the evaluation, avoids the interference of human factors, and improves the evaluation efficiency because the key indicators and evaluation criteria have been pre-set and stored in the database and can be directly called.

[0168] S402. Establish a supplier database and collect basic information of suppliers.

[0169] Specifically, the control system establishes a supplier database and collects basic information of suppliers. The basic information of suppliers includes company qualifications, scale, experience, and performance. A dedicated supplier database is established to store the basic information of suppliers. The basic information of suppliers is collected through various channels, including company qualifications (such as business license, quality management system certification, etc.), scale (such as number of employees, annual output value, etc.), experience (such as years of operation in the industry, successful cases, etc.), and performance (such as historical supply situation, customer satisfaction, etc.). The collected information is sorted and classified to ensure the accuracy and integrity of the information. A comprehensive supplier information database is established, providing data support for subsequent evaluations and selections. It improves the efficiency of information management and facilitates the querying and updating of supplier information.

[0170] S403. Conduct a multi-dimensional comprehensive evaluation of suppliers and regularly conduct performance evaluations of suppliers.

[0171] Specifically, the control system conducts a multi-dimensional comprehensive evaluation of suppliers, regularly evaluates the performance of suppliers, and selects suppliers with high stability and reliability; according to the preset evaluation criteria and key indicators, a multi-dimensional comprehensive evaluation of suppliers is carried out. The evaluation may include quality evaluation (such as product qualification rate, defective rate, etc.), delivery evaluation (such as delivery punctuality, delivery cycle, etc.), price evaluation (such as price competitiveness, cost savings, etc.), and service evaluation (such as technical support, after-sales service, etc.). A scoring system or a grading system is used to quantitatively evaluate suppliers, so as to more intuitively compare the performance of different suppliers. It ensures the comprehensiveness and objectivity of the evaluation, can more accurately reflect the actual performance of suppliers, and provides a scientific basis for subsequent supplier selection and performance evaluation.

[0172] Set a regular performance evaluation cycle, such as once every quarter or once a year. During the evaluation cycle, according to the actual performance of the supplier, combined with the preset evaluation criteria and key indicators, the performance of the supplier is evaluated. The evaluation results are summarized and analyzed to find out the advantages and disadvantages of the supplier and put forward improvement suggestions. It promotes the continuous improvement and enhancement of suppliers, improves the stability and reliability of the supply chain, and provides a basis for the enterprise to select suppliers with high stability and reliability.

[0173] S404. Regularly review and adjust the evaluation indicators and criteria.

[0174] Specifically, the control system regularly reviews and adjusts the evaluation indicators and criteria to ensure that they always meet the actual needs of the enterprise. According to market changes and changes in enterprise needs, the evaluation indicators and criteria are regularly reviewed and adjusted. The review may include an analysis of the effectiveness, applicability, etc. of the evaluation indicators; the adjustment includes adding new evaluation indicators, modifying the original evaluation criteria, etc. It ensures that the evaluation indicators and criteria keep pace with the times and always meet the actual needs of the enterprise, improves the accuracy and effectiveness of the evaluation, and provides strong support for the enterprise to select high-quality suppliers.

[0175] Based on the above method, the embodiment of the present application also discloses a loading and distribution control system for electric power metering materials. A loading and distribution control system for electric power metering materials includes:

[0176] A real-time usage data generation module, which is used to track and monitor materials through the Internet of Things, RFID technology, and barcode technology, and generate real-time usage data;

[0177] A historical monthly demand data query module, which is used to query historical monthly demand data from a preset database;

[0178] An equipment demand prediction module, which is used to predict the future equipment demand according to the historical monthly demand data and real-time usage data by using time series analysis and machine learning methods;

[0179] A probability distribution function generation module is configured to mine the probability distribution characteristics of the power metering material requirements on the basis of deterministic prediction, so as to generate a probability distribution function, which is used to help users configure the inventory more flexibly and avoid the problem of unreasonable inventory caused by untrue demand information.

[0180] An embodiment of the present application further discloses an intelligent terminal, which includes a memory and a processor. The memory stores a computer program that can be loaded and executed by the processor, such as a loading and distribution control method for a power metering material as described above.

[0181] An embodiment of the present application further discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program that can be loaded and executed by the processor, such as a loading and distribution control method for a power metering material as described above. The computer-readable storage medium includes, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.

[0182] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the protection scope of the invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art can still, without conflict, make combinations, additions, deletions, or other adjustments to the features in the embodiments of the present invention according to the situation without creative efforts, so as to obtain different technical solutions that are essentially not deviated from the concept of the present invention. These technical solutions also belong to the scope of protection of the present invention.

Claims

1. A method for controlling the loading and distribution of electric power metering materials, characterized in that: The following steps are involved: Track and monitor materials through the Internet of Things, RFID technology, and barcode technology to generate real-time usage data; Query historical monthly demand data from the preset database; Based on the historical monthly demand data and real-time usage data, use time series analysis and machine learning methods to predict future equipment demand; On the basis of deterministic prediction, the probability distribution characteristics of the demand for electricity metering materials are mined to generate a probability distribution function to help users configure inventory more flexibly and avoid unreasonable inventory problems caused by untrue demand information.

2. A method for controlling the loading and distribution of electric power metering materials according to claim 1, characterized in that: The steps of using time series analysis and machine learning methods to predict future equipment demand specifically include: Pre-process the historical monthly demand data, including removing abnormal data and normalizing data to ensure data quality and consistency; Collect and integrate influencing factor data, including weather, equipment output, attributes, and grid dispatch data; Use time series analysis models to analyze trends, cycles, and noise components in historical monthly demand data and predict future demand; Construct the first demand forecasting model based on time series to reflect the impact of time factors on demand; Constructing a neural network model, and using the historical monthly demand data to train the neural network model to capture the dependency between power demand and different characteristics; Use LSTM networks to forecast monthly, quarterly, and annual demand, and improve forecast accuracy by weighted averaging monthly, quarterly, and annual forecast results; Combined with machine learning algorithms, it broadens the modeling capabilities of traditional time series algorithms; Based on the time series multi-factor fusion model, different influencing factors are incorporated into the forecasting model to obtain the second demand forecasting model to improve the comprehensiveness and accuracy of the forecast; Combining the first demand forecasting model and the second demand forecasting model through the maximum information entropy principle to form a comprehensive forecasting model; In practical applications, a comprehensive prediction model is deployed and tested to compare the prediction results with actual needs. The effectiveness of the time series prediction model is verified through examples, and the model is adjusted according to actual needs.

3. The method for controlling the loading and distribution of electric power metering materials according to claim 1 is characterized in that: The step of mining the probability distribution characteristics of the demand for electric energy metering materials on the basis of deterministic prediction to generate a probability distribution function also includes: Collect and organize historical demand data of various types of electric energy metering materials, including actual demand quantity per month, as well as the frequency and periodic characteristics of material usage; Based on the historical demand data, a demand forecasting model is established using time series analysis, machine learning models or statistical methods to capture trends and patterns in the historical demand data and predict future demand; Performing statistical analysis on the historical demand data, calculating mean and variance parameters, and using probability distribution functions to describe the probability characteristics of power material demand, and calculating the probability distribution function of material demand in the future through a prediction model; Query predetermined confidence levels and historical data from a pre-set database; Using the generated probability distribution function, calculate the safety stock quantity under the predetermined confidence level; Set safety stock levels based on the probability distribution of demand to cope with the risk of demand fluctuations; Set inventory strategy parameters based on historical data and business needs; Use simulation software or programming tools to simulate the operation of the inventory system and record the changes in key indicators during the simulation; Analyze the service level and cost-effectiveness under different inventory strategies and select the optimal strategy; The CRPS indicator is used to evaluate the accuracy and reliability of probability predictions, the model is calibrated to ensure the consistency of the prediction results with the actual observations, and the prediction performance is optimized by adjusting the model parameters or using different kernel functions.

4. The method for controlling the loading and distribution of electric power metering materials according to claim 1 is characterized in that: It also includes the use of information systems to achieve intelligent distribution route planning for power metering materials, including: By integrating the Internet of Things, big data, cloud computing and artificial intelligence technologies, a comprehensive intelligent scheduling platform is established to achieve real-time matching of material demand and supply, optimize the matching of vehicles and distribution routes, and improve overall operational efficiency; In terms of route planning, a combination of genetic algorithm and hybrid particle swarm optimization algorithm is used to improve computational efficiency and reduce the probability of local optimal solutions, thereby obtaining the best delivery route. Use heuristic orthogonal binary tree search algorithm to reasonably configure the measurement material vehicles, and formulate distribution plans based on actual needs to optimize distribution routes and loading efficiency; Through the Geographic Information System (GIS) and the Global Positioning System (GPS), the location of the delivery vehicles is monitored in real time to optimize the location and layout of the delivery center; the GIS is also used to construct a geospatial data preprocessing model to provide a scientific basis for route planning; Develop an information-based material distribution system based on the ERP system to achieve the integration of warehouse management, transportation management and data analysis management; automatically generate logistics orders through data interface docking with the ERP system to achieve intelligent distribution management.

5. A method for controlling the loading and distribution of electric power metering materials according to claim 4, characterized in that: In terms of path planning, the steps of improving the computational efficiency by combining the genetic algorithm with the hybrid particle swarm optimization algorithm specifically include: Use the triangle method to model the spatial environment and generate basic data for initial path planning; Implement the artificial potential field method in the genetic algorithm and design the initial path planning; The target point is set as the attraction point, the obstacle is set as the repulsion point, and the moving object is guided to approach the target position through the virtual force field; A parallel search method is used to find the global optimal solution, combining the genetic algorithm with the artificial potential field method, and the artificial potential field method is used to optimize the global path obtained by the genetic algorithm; Combining the advantages of genetic algorithm and particle swarm optimization algorithm, a hybrid algorithm is formed; the initial path generated by the genetic algorithm is used as the input of the particle swarm optimization algorithm, and the path is further optimized by the particle swarm optimization algorithm; the particle swarm optimization algorithm adjusts the position and speed of particles and uses global search capabilities and local search capabilities to quickly converge to the optimal solution; The path generated by the hybrid algorithm is smoothed using the cubic spline interpolation method to reduce the risk of collision in the path; The effectiveness of the proposed algorithm is verified through simulation experiments.

6. A method for controlling the loading and distribution of electric power metering materials according to claim 3, characterized in that: It also includes establishing a material demand plan and a safety stock quota system, and optimizing procurement strategies through data analysis to reduce inventory backlogs and capital occupation, including: Apply machine learning models to conduct in-depth analysis of material usage, identify demand fluctuations, seasonal factors, and market trends, and develop demand forecasting models; Through big data analysis technology, historical sales data, market trends and consumer behavior are analyzed to predict future material demand; Based on the results of big data analysis, develop an intelligent decision support system to automatically generate material demand plans, procurement recommendations, and inventory adjustment plans based on preset rules and algorithms; Monitor inventory levels and sales velocity in real time, identify risks of overstocking or out-of-stock, and adjust replenishment strategies based on real-time data to ensure sufficient inventory without waste; According to the results of big data analysis, the safety stock level is adjusted to reduce inventory costs and lower capital occupation; by analyzing inventory turnover rate and out-of-stock rate indicators, unsalable, overstocked and short-supply materials are identified, and corresponding adjustment strategies are formulated.

7. A method for controlling the loading and distribution of electric power metering materials according to claim 6, characterized in that: Also includes: Query key indicators and evaluation criteria from a preset database, where the key indicators include supplier quality, delivery punctuality, price competitiveness, and supply capacity; the evaluation criteria are pre-set by the user based on the specific needs and characteristics of the enterprise; Establish a supplier database and collect basic information about suppliers, including company qualifications, scale, experience, and performance; Conduct comprehensive evaluation of suppliers from multiple dimensions, regularly evaluate supplier performance, and select suppliers with high stability and reliability; Regularly review and adjust evaluation indicators and standards to ensure they always meet the actual needs of the enterprise.

8. A loading and distribution control system for electric power metering materials, characterized in that: include: Real-time usage data generation module, used to track and monitor materials through the Internet of Things, RFID technology, and barcode technology to generate real-time usage data; A historical monthly demand data query module is used to query historical monthly demand data from a preset database; An equipment demand forecasting module, used to forecast future equipment demand using time series analysis and machine learning methods based on the historical monthly demand data and real-time usage data; The probability distribution function generation module is used to mine the probability distribution characteristics of the demand for electricity metering materials on the basis of deterministic prediction, so as to generate a probability distribution function to help users configure inventory more flexibly and avoid unreasonable inventory problems caused by false demand information.

Images